[x265] [PATCH] blockcopy_pp_2x4: SSE2 asm code optimization

chen chenm003 at 163.com
Thu Feb 5 13:17:06 CET 2015


Why?
old code are less size.

At 2015-02-05 18:35:54,praveen at multicorewareinc.com wrote:
># HG changeset patch
># User Praveen Tiwari
># Date 1423132546 -19800
># Node ID 18edf28d3879454f940fced1e8f5b6cb81fe041f
># Parent  1265aafe5af1d66b8151ffd9bdc6fe595f7e6343
>blockcopy_pp_2x4: SSE2 asm code optimization
>
>diff -r 1265aafe5af1 -r 18edf28d3879 source/common/x86/blockcopy8.asm
>--- a/source/common/x86/blockcopy8.asm Thu Feb 05 14:54:55 2015 +0530
>+++ b/source/common/x86/blockcopy8.asm Thu Feb 05 16:05:46 2015 +0530
>@@ -47,15 +47,15 @@
> cglobal blockcopy_pp_2x4, 4, 7, 0
>     mov    r4w,    [r2]
>     mov    r5w,    [r2 + r3]
>-    lea    r2,     [r2 + r3 * 2]
>-    mov    r6w,    [r2]
>+    mov    r6w,    [r2 + 2 * r3]
>+    lea    r3,     [r3 + 2 * r3]
>     mov    r3w,    [r2 + r3]
> 
>-    mov    [r0],         r4w
>-    mov    [r0 + r1],    r5w
>-    lea    r0,           [r0 + 2 * r1]
>-    mov    [r0],         r6w
>-    mov    [r0 + r1],    r3w
>+    mov    [r0],          r4w
>+    mov    [r0 + r1],     r5w
>+    mov    [r0 + 2 * r1], r6w
>+    lea    r1,            [r1 + 2 * r1]
>+    mov    [r0 + r1],     r3w
> RET
> 
> ;-----------------------------------------------------------------------------
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150205/4495ff80/attachment.html>


More information about the x265-devel mailing list